No. 420.] AN EXTRAORDINARY ANT-GUEST. IOI5 
First, the peculiar habits of the phorid show clearly that the 
ponerine method of feeding the larvae with comminuted insects 
is not only the typical but the only method employed by these 
ants, for such a commensal would certainly starve if the Pachy- 
condyla larve were carefully fed like the larvae of Camponotus 
and Formica, by regurgitation of liquid food from the mouths 
of the workers. The phorid profits by a peculiarity in the 
behavior of its host, and thereby demonstrates — by one of 
Nature's experiments — that Pachycondyla harpax cannot feed 
its young by regurgitation. 
Second, in the peculiar symbiotic relationship existing 
between the phorid and the ants, the adaptations are all on 
the part of the former, whereas the latter pursues its ancient 
and well-established mode of life uninfluenced by and apparently 
in complete ignorance of the very existence of its little guests. 
Even the distortion of the cocoon may be due, as I have sug- 
gested, entirely to the activity of the phorid. These strikingly 
unilateral adaptations are probably to be explained on the - 
ground that the phorid in the Pachycondyla nest is as careful 
and conservative — if such terms may be used — of the life 
and welfare of its host as Pergande's Apocephalus is ruth- 
less and destructive, The small amount of food consumed by 
the former can hardly be a serious drain on the provisioning 
instincts of the Ponerine. The larvae bearing the commensals 
were certainly as large and healthy as any others in the nest, 
and produced perfectly normal pupa, which in the cases 
observed all lacked the imaginal disks for the wings and were 
therefore of the worker type. Whether the presence of the 
commensals by reducing the amount of food even to a slight 
degree could inhibit the development of queen larva and thereby 
convert them into workers, involves a problem as interesting 
as it is difficult to solve in the present state of our knowledge. 
Third, the commensal of Pachycondyla, together with the 
Phoridæ mentioned in the introductory paragraph of this 
paper, shows that the flies of this family are well MORE care- 
ful study. While the Stratiomyide, Syrphidze, Midaide, etc., 
certainly make more attractive cabinet specimens and pret- 
tier monographic plates, it is very doubtful whether the life 
